Control Systems Investment and Return Part 2 of 3 By F.G. Shinskey Sponsored by: 2012 ExperTune 1
Objective CONTROL SYSTEMS: INVESTMENT AND RETURN with Examples from Industry F. G. Shinskey Process Control Consultant Part 2 of a 3-part series Part 2 reviews Control System Objectives and shares case studies that demonstrate the benefits of advanced controls. Additional case studies demonstrate how control system improvements can increase production, and help to ensure regulatory compliance. The objective of a control system is to regulate process variables in such a way as to operate the plant at the safest, most productive and profitable conditions possible consistent with the existing market, available feedstock, and equipment limitations automatically that is, with minimal human assistance. There are two basic obstacles to reaching this goal in any given process: Instability and Ineffectiveness. Instability is the result of control overreaction to a variation in the controlled variable, inducing a later and opposite variation of similar magnitude. It appears as cycling in one or more controlled variables, whose amplitude may be constant or not. A constantamplitude cycle, known as a limit cycle, is caused by a nonlinear element in the control loop. Limit cycles tend to be non-sinusoidal, and are not correctible by controller tuning or replacement the nonlinear element (such as valve deadband) must be overcome. If the loop is linear, its sinusoidal cycle may either expand or contract, but an expanding cycle will eventually be limited by some physical stop, such as valve travel. Instability is costly in terms of wear on components, reduced production capacity, and inefficient utilization of resources such as energy. Ineffectiveness is the inability to maintain the controlled variable at set point in the presence of variations in ambient conditions, feed rate and composition, and upsets caused by the actions of other controllers. The worst case is no control at all; next is Manual control that receives occasional operator attention; other cases include poorly tuned controllers, mismatched valve characteristics, and incorrectly configured loops. Control ineffectiveness also wastes resources, limits production capacity, and gives away valuable product. Examples follow with corrections that saved dollars lots of dollars. 2
Case 4: Model-predictive Control (MPC) is being used to optimize many processes, including distillation. But its success rate has not been high, in many cases because the process itself cannot be stabilized, and stability is essential before optimization can begin. The MPC typically manipulates temperature set points one in the upper section of the column, and the other in the lower section. If the temperature controllers (TC) in turn manipulate reflux and boilup flows the most common arrangement their level of interaction is likely to be high. Increasing reflux lowers both temperatures and increasing boilup raises both, in very similar degrees. Loop interaction can be measured by Relative Gain Analysis (RGA) where a value of 1.00 is ideal and interaction worsens in either direction away from 1.00. An MPC system was being applied to a debutanizer column in this way, where RGA calculations indicated a value of 12. Not surprisingly, the two temperatures cycled against one another endlessly and the MPC system could not overcome it. Not only was optimization impossible, but cycling also uses more energy than operation at the same average flows and compositions in the steady state. The top temperature loop was eventually reconfigured to manipulate reflux/distillate ratio rather than reflux flow, which dropped the RG to 2. Immediately the cycling stopped, and optimization could begin. It was even possible to lower the proportional band of the lower TC by a factor of 6 the RG correction factor. This transformation is described in Ref. 6. 3
Cases 5 & 6: Maximize Polymer Production Batch reactors have historically been used to produce polystyrene, several hours being required to make a batch. A critical point in the process involves heating the reaction mass to the desired temperature for the reaction to proceed as quickly as possible, but without overshoot. A fast ascent increases productivity, but a severe overshoot can ruin a batch and require cleaning of the vessel at a high cost. Integral windup in the TC causes severe overshoot, if not disabled. A dual-mode control system was developed to raise temperature as fast as possible with full heating, then switch to full cooling, then to preloaded PID control. The switching point was only about 2-3 F below set point, and the delay during full cooling less than a minute, so startup time was as short as possible. Preloading the integral term with the anticipated steady-state controller output was necessary to minimize any bump when transferring to PID control. The system was successful from the very first batch, eliminating losses and minimizing run time. Production was maximized and operators life made easier. The system is described in its first implementation using pneumatic components in Ref. 7. In later digital systems, the trip point and the preload setting were adapted according to the observed rate of temperature rise. A polypropylene plant had several continuous reactors, one of which was limit-cycling when attempts were made to raise its temperature to the desired set point for production. Reactor temperature had been controlled satisfactorily in the past, and other similar reactors at the plant were being controlled successfully at the same time. A cascade system was being used, where the primary TC manipulated the set point of the cooler inlet temperature in cascade. Attempts to retune the primary TC had no effect on the sawtooth cycle. Moving the secondary TC from the inlet of the cooler to the outlet produced the same results. An adaptive controller was even tried, unsuccessfully. Persistent limit-cycling can t be eliminated by retuning or reconfiguring control loops something was wrong with the reactor. Exothermic reactors can be stable, unstable, or uncontrollable as described in Ref. 8. A stable reactor can be controlled manually, but an unstable reactor will run away without control. The difference between them is the degree of self-regulation built into the heattransfer system in this case the aforementioned cooler. An unstable stirred-tank reactor is controllable because its lag-dominance (low dynamic gain) allows a high controller gain. This reactor was definitely unstable, and being a tubular device dominated by dead time, was also uncontrollable. The good news was that it had been stable in the past, as were the others in the plant. So the problem was determined to be a fouled heat-transfer surface in the cooler. After the cooler was cleaned and returned to service, reactor temperature was controlled within ±0.1 F, without any changes to the control system or its tuning, and full production restored. 4
Case 7: Insure Compliance in Wastewater Treatment If effluent ph from an industrial facility fails to meet local regulatory standards, the plant can be fined, and repeat offenders shutdown. Successful ph control depends on many factors: vessel layout and mixing, reagent piping and valving, holding tanks for impounding off-spec waste and recycling it, are all important. Electrodes must be properly positioned and kept clean of fouling substances that can increase measurement lag. Because the relationship between ph and reagent delivery is fundamentally logarithmic, nonlinear characterization is often required in the feedback controller. And where the wastewater contains a variety of ionic species from strong to weak agents, that characterization may have to be adapted on-line to keep the control loop both responsive and stable. Reference 9 describes an installation where the wastewater from diverse production lines in a chemical plant changed from weak to strong randomly, so that manual tuning of the ph controller was not sufficient to keep the effluent within allowable limits. An adaptive function was added that detected whether the ph was cycling or drifting away from set point, and loosened or tightened the controller s nonlinear function accordingly. Not only was the system completely effective in preventing off-specification discharges, but the elimination of prolonged cycling cut reagent usage in half. 5
References in Parts 1-3 1. Shinskey, F.G., PID-deadtime Control of Distributed Processes, Control Eng. Practice, 9 (2001) 1177-1183. 2. Shinskey, F.G., and J.R. Louis, Once-through Boiler Control System, U.S. Patent 3,417,737, Dec. 24, 1968. 3. Shinskey, F.G., Smoothing Out Compressor Control, Chem. Eng., Feb. 1999, pp. 127-130. 4. Shinskey, F.G., Minimizing Operating Costs for Distillation Columns, Oil & Gas J., July 21, 1969, pp. 79-82. 5. Shinskey, F.G., Process Control Systems, 4 th ed., McGraw-Hill, New York, (1996) p. 361. 6. Shinskey, F.G., Multivariable Control of Distillation, Control, May-July, 2009. 7. Shinskey, F.G. and J.L. Weinstein, A Dual-mode Control System for a Batch Exothermic Reactor, 20 th Annual ISA Conference, October 1965. 8. Shinskey, F.G., Exothermic Reactors: the Stable, the Unstable, and the Uncontrollable, Chem. Eng., March 2004, pp. 54-58. 9. Shinskey, F.G., Adaptive Nonlinear Control System, U.S. Patent 3,794,817, Feb. 26, 1974. 10. Shinskey, F.G., Taming the Shrink-Swell Dragon, Control, March 2004. 11. Shinskey, F.G., How to Control Product Dryness without measuring it. InTech, Sept. 1968, pp. 47-51. 12. Shinskey, F.G., Batch Dryer Control Apparatus, U.S. Patent 3.699,665, Oct. 24, 1972. 13. Shinskey, F.G., Flow and Pressure Control Using Variable-speed Motors, Contr. Eng. Conference, Chicago, May 1982. 14. Fauth, C.F., and F.G. Shinskey, Advanced Control of Distillation Columns, Chem. Eng. Progr., June 1975, pp. 49-54. 6